Hugging Face's logo

tomaarsen
/
bert-base-uncased-tsdae-askubuntu

Sentence Similarity
sentence-transformers
Safetensors
bert
feature-extraction
Generated from Trainer
dataset_size:160436
loss:DenoisingAutoEncoderLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
bert-base-uncased-tsdae-askubuntu / README.md
tomaarsen's picture
tomaarsen HF staff
Add new SentenceTransformer model
cc6732f verified
preview code
|
raw
history blame contribute delete
16 kB
metadata
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:160436
  - loss:DenoisingAutoEncoderLoss
base_model: google-bert/bert-base-uncased
widget:
  - source_sentence: >-
      how do i make evolution check and notify new emails , without keeping main
      ui open ?
    sentences:
      - ppas be removed?
      - >-
        how set serve as a samba primary controller pam modules to authenticate
        against?
      - how do make check and notify new emails keeping
  - source_sentence: setting http proxy in awesome wm
    sentences:
      - on 10.04 on p series?
      - setting http proxy awesome wm
      - mean package is "set to installed?
  - source_sentence: what is ubuntu advantage ?
    sentences:
      - is advantage?
      - how turn calling on f1
      - is utnubu?
  - source_sentence: is there a way to check hardware integrity ?
    sentences:
      - is there a way to hardware integrity?
      - to change key ctrl
      - software is to tv card
  - source_sentence: how to fix ssl error from python apps ( urllib ) when behind https proxy ?
    sentences:
      - windows started with archive
      - upstart
      - how to ssl from python () proxy
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - map
  - mrr@10
  - ndcg@10
co2_eq_emissions:
  emissions: 74.02946721860093
  energy_consumed: 0.19045301341027557
  source: codecarbon
  training_type: fine-tuning
  on_cloud: false
  cpu_model: 13th Gen Intel(R) Core(TM) i7-13700K
  ram_total_size: 31.777088165283203
  hours_used: 0.64
  hardware_used: 1 x NVIDIA GeForce RTX 3090
model-index:
  - name: SentenceTransformer based on google-bert/bert-base-uncased
    results:
      - task:
          type: reranking
          name: Reranking
        dataset:
          name: AskUbuntu dev
          type: AskUbuntu-dev
        metrics:
          - type: map
            value: 0.5058158414596666
            name: Map
          - type: mrr@10
            value: 0.6325571254142682
            name: Mrr@10
          - type: ndcg@10
            value: 0.5529143206799554
            name: Ndcg@10
      - task:
          type: reranking
          name: Reranking
        dataset:
          name: AskUbuntu test
          type: AskUbuntu-test
        metrics:
          - type: map
            value: 0.5826205294809574
            name: Map
          - type: mrr@10
            value: 0.7237319322514852
            name: Mrr@10
          - type: ndcg@10
            value: 0.6303658219971641
            name: Ndcg@10

SentenceTransformer based on google-bert/bert-base-uncased

This is a sentence-transformers model finetuned from google-bert/bert-base-uncased. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

  • Model Type: Sentence Transformer
  • Base model: google-bert/bert-base-uncased
  • Maximum Sequence Length: 75 tokens
  • Output Dimensionality: 768 dimensions
  • Similarity Function: Cosine Similarity

Model Sources

Full Model Architecture 

SentenceTransformer(
  (0): Transformer({'max_seq_length': 75, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("tomaarsen/bert-base-uncased-tsdae-askubuntu")
# Run inference
sentences = [
    'how to fix ssl error from python apps ( urllib ) when behind https proxy ?',
    'how to ssl from python () proxy',
    'upstart',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Reranking

Metric AskUbuntu-dev AskUbuntu-test
map 0.5058 0.5826
mrr@10 0.6326 0.7237
ndcg@10 0.5529 0.6304

Training Details

Training Dataset

Unnamed Dataset

  • Size: 160,436 training samples
  • Columns: text and noisy
  • Approximate statistics based on the first 1000 samples:
    text noisy
    type string string
    details
    • min: 5 tokens
    • mean: 14.43 tokens
    • max: 39 tokens
    • min: 3 tokens
    • mean: 9.47 tokens
    • max: 24 tokens
  • Samples:
    text noisy
    how to get the `` your battery is broken '' message to go away ? to get the is broken go away?
    how can i set the software center to install software for non-root users ? how can i the center install non-root users
    what are some alternatives to upgrading without using the standard upgrade system ? what are alternatives to using standard system?
  • Loss: DenoisingAutoEncoderLoss

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • learning_rate: 3e-05
  • num_train_epochs: 1
  • warmup_ratio: 0.1
  • fp16: True

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 8
  • per_device_eval_batch_size: 8
  • per_gpu_train_batch_size: None
  • per_gpu_eval_batch_size: None
  • gradient_accumulation_steps: 1
  • eval_accumulation_steps: None
  • torch_empty_cache_steps: None
  • learning_rate: 3e-05
  • weight_decay: 0.0
  • adam_beta1: 0.9
  • adam_beta2: 0.999
  • adam_epsilon: 1e-08
  • max_grad_norm: 1.0
  • num_train_epochs: 1
  • max_steps: -1
  • lr_scheduler_type: linear
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.1
  • warmup_steps: 0
  • log_level: passive
  • log_level_replica: warning
  • log_on_each_node: True
  • logging_nan_inf_filter: True
  • save_safetensors: True
  • save_on_each_node: False
  • save_only_model: False
  • restore_callback_states_from_checkpoint: False
  • no_cuda: False
  • use_cpu: False
  • use_mps_device: False
  • seed: 42
  • data_seed: None
  • jit_mode_eval: False
  • use_ipex: False
  • bf16: False
  • fp16: True
  • fp16_opt_level: O1
  • half_precision_backend: auto
  • bf16_full_eval: False
  • fp16_full_eval: False
  • tf32: None
  • local_rank: 0
  • ddp_backend: None
  • tpu_num_cores: None
  • tpu_metrics_debug: False
  • debug: []
  • dataloader_drop_last: False
  • dataloader_num_workers: 0
  • dataloader_prefetch_factor: None
  • past_index: -1
  • disable_tqdm: False
  • remove_unused_columns: True
  • label_names: None
  • load_best_model_at_end: False
  • ignore_data_skip: False
  • fsdp: []
  • fsdp_min_num_params: 0
  • fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
  • fsdp_transformer_layer_cls_to_wrap: None
  • accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
  • deepspeed: None
  • label_smoothing_factor: 0.0
  • optim: adamw_torch
  • optim_args: None
  • adafactor: False
  • group_by_length: False
  • length_column_name: length
  • ddp_find_unused_parameters: None
  • ddp_bucket_cap_mb: None
  • ddp_broadcast_buffers: False
  • dataloader_pin_memory: True
  • dataloader_persistent_workers: False
  • skip_memory_metrics: True
  • use_legacy_prediction_loop: False
  • push_to_hub: False
  • resume_from_checkpoint: None
  • hub_model_id: None
  • hub_strategy: every_save
  • hub_private_repo: None
  • hub_always_push: False
  • gradient_checkpointing: False
  • gradient_checkpointing_kwargs: None
  • include_inputs_for_metrics: False
  • include_for_metrics: []
  • eval_do_concat_batches: True
  • fp16_backend: auto
  • push_to_hub_model_id: None
  • push_to_hub_organization: None
  • mp_parameters:
  • auto_find_batch_size: False
  • full_determinism: False
  • torchdynamo: None
  • ray_scope: last
  • ddp_timeout: 1800
  • torch_compile: False
  • torch_compile_backend: None
  • torch_compile_mode: None
  • dispatch_batches: None
  • split_batches: None
  • include_tokens_per_second: False
  • include_num_input_tokens_seen: False
  • neftune_noise_alpha: None
  • optim_target_modules: None
  • batch_eval_metrics: False
  • eval_on_start: False
  • use_liger_kernel: False
  • eval_use_gather_object: False
  • average_tokens_across_devices: False
  • prompts: None
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss AskUbuntu-dev_map AskUbuntu-test_map
-1 -1 - 0.4151 -
0.0499 1000 6.1757 - -
0.0997 2000 4.0925 - -
0.1496 3000 3.2921 - -
0.1995 4000 2.9046 - -
0.2493 5000 2.669 0.5158 -
0.2992 6000 2.5884 - -
0.3490 7000 2.437 - -
0.3989 8000 2.3406 - -
0.4488 9000 2.2709 - -
0.4986 10000 2.1881 0.5131 -
0.5485 11000 2.1627 - -
0.5984 12000 2.1055 - -
0.6482 13000 2.0577 - -
0.6981 14000 2.0133 - -
0.7479 15000 1.9877 0.5130 -
0.7978 16000 1.9569 - -
0.8477 17000 1.9219 - -
0.8975 18000 1.9124 - -
0.9474 19000 1.8676 - -
0.9973 20000 1.8461 0.5058 -
-1 -1 - - 0.5826

Environmental Impact

Carbon emissions were measured using CodeCarbon.

  • Energy Consumed: 0.190 kWh
  • Carbon Emitted: 0.074 kg of CO2
  • Hours Used: 0.64 hours

Training Hardware

  • On Cloud: No
  • GPU Model: 1 x NVIDIA GeForce RTX 3090
  • CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
  • RAM Size: 31.78 GB

Framework Versions

  • Python: 3.11.6
  • Sentence Transformers: 3.4.0.dev0
  • Transformers: 4.48.0.dev0
  • PyTorch: 2.5.0+cu121
  • Accelerate: 0.35.0.dev0
  • Datasets: 2.20.0
  • Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers 

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

DenoisingAutoEncoderLoss 

@inproceedings{wang-2021-TSDAE,
    title = "TSDAE: Using Transformer-based Sequential Denoising Auto-Encoderfor Unsupervised Sentence Embedding Learning",
    author = "Wang, Kexin and Reimers, Nils and Gurevych, Iryna",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
    month = nov,
    year = "2021",
    address = "Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    pages = "671--688",
    url = "https://arxiv.org/abs/2104.06979",
}